initial commit

README.md

@@ -1,3 +0,0 @@
----
-license: mit
----

config.json

@@ -0,0 +1,39 @@
+{
+  "_name_or_path": "gpt2",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.12.0.dev0",
+  "use_cache": true,
+  "vocab_size": 50257
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87fceebaf493fcdf5a577b7458b572165e44ec01a4ec90ae484f1827fb46b3dd
+size 664786896

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e0cd50bb26e2e012060b2b9268e8803dbe6e714ce77f04d184c003b8ef7e998
+size 13620

rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0ef458b36cb65b3573443b4dc3328bac773ad28103f5b4a8871708381a87a60
+size 13619

rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3701129e1336bc63d908a898b505e402dbca873258774f660cfafed89bdf258e
+size 13620

rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fc47ae2f0260e48fda5d85b5f7226a7994cde5150dde12784f43cee2d4308f5
+size 13619

rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af5e376370ba0d1ae1d22a73f898df61300d81f3220b8044e14f4ca9521bce34
+size 13620

rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:328031933d3e78bc9d41125c8207051b003e0957b7b9d093ccdc0c29ceb279b8
+size 13619

rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51075f868424d14d859d50d63b8dec44ca43b439272e0e05d8efbad8e9a75cae
+size 13620

rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4602385985ceb63a91a629abf5a3bb8372d74d10602c816e4dad3d8cf2ebc2bf
+size 13619

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79bcfb2d6cca4c63cdfcd719830fc3cf7df95bc9e58af991b6fdeff67b97f21c
+size 623

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "models/gpt2_uyghur", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -0,0 +1,317 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 131.57894736842104,
+  "global_step": 17500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 3.76,
+      "learning_rate": 1e-05,
+      "learning_rate_embeddings": 1e-05,
+      "loss": 9.5582,
+      "step": 500
+    },
+    {
+      "epoch": 7.52,
+      "learning_rate": 2e-05,
+      "learning_rate_embeddings": 2e-05,
+      "loss": 8.0322,
+      "step": 1000
+    },
+    {
+      "epoch": 11.28,
+      "learning_rate": 3e-05,
+      "learning_rate_embeddings": 3e-05,
+      "loss": 7.3149,
+      "step": 1500
+    },
+    {
+      "epoch": 15.04,
+      "learning_rate": 4e-05,
+      "learning_rate_embeddings": 4e-05,
+      "loss": 6.6378,
+      "step": 2000
+    },
+    {
+      "epoch": 18.8,
+      "learning_rate": 5e-05,
+      "learning_rate_embeddings": 5e-05,
+      "loss": 6.0786,
+      "step": 2500
+    },
+    {
+      "epoch": 18.8,
+      "eval_loss": 5.710835933685303,
+      "eval_runtime": 23.8206,
+      "eval_samples_per_second": 315.399,
+      "eval_steps_per_second": 2.477,
+      "step": 2500
+    },
+    {
+      "epoch": 22.56,
+      "learning_rate": 6e-05,
+      "learning_rate_embeddings": 6e-05,
+      "loss": 5.6654,
+      "step": 3000
+    },
+    {
+      "epoch": 26.32,
+      "learning_rate": 7.000000000000001e-05,
+      "learning_rate_embeddings": 7.000000000000001e-05,
+      "loss": 5.3417,
+      "step": 3500
+    },
+    {
+      "epoch": 30.08,
+      "learning_rate": 8e-05,
+      "learning_rate_embeddings": 8e-05,
+      "loss": 5.0624,
+      "step": 4000
+    },
+    {
+      "epoch": 33.83,
+      "learning_rate": 8.999999999999999e-05,
+      "learning_rate_embeddings": 8.999999999999999e-05,
+      "loss": 4.8364,
+      "step": 4500
+    },
+    {
+      "epoch": 37.59,
+      "learning_rate": 0.0001,
+      "learning_rate_embeddings": 0.0001,
+      "loss": 4.6372,
+      "step": 5000
+    },
+    {
+      "epoch": 37.59,
+      "eval_loss": 4.484490394592285,
+      "eval_runtime": 10.1862,
+      "eval_samples_per_second": 737.569,
+      "eval_steps_per_second": 5.792,
+      "step": 5000
+    },
+    {
+      "epoch": 41.35,
+      "learning_rate": 0.00011,
+      "learning_rate_embeddings": 0.00011,
+      "loss": 4.4734,
+      "step": 5500
+    },
+    {
+      "epoch": 45.11,
+      "learning_rate": 0.00012,
+      "learning_rate_embeddings": 0.00012,
+      "loss": 4.322,
+      "step": 6000
+    },
+    {
+      "epoch": 48.87,
+      "learning_rate": 0.00013000000000000002,
+      "learning_rate_embeddings": 0.00013000000000000002,
+      "loss": 4.1937,
+      "step": 6500
+    },
+    {
+      "epoch": 52.63,
+      "learning_rate": 0.00014000000000000001,
+      "learning_rate_embeddings": 0.00014000000000000001,
+      "loss": 4.0804,
+      "step": 7000
+    },
+    {
+      "epoch": 56.39,
+      "learning_rate": 0.00015,
+      "learning_rate_embeddings": 0.00015,
+      "loss": 3.9741,
+      "step": 7500
+    },
+    {
+      "epoch": 56.39,
+      "eval_loss": 3.9806618690490723,
+      "eval_runtime": 10.0742,
+      "eval_samples_per_second": 745.765,
+      "eval_steps_per_second": 5.857,
+      "step": 7500
+    },
+    {
+      "epoch": 60.15,
+      "learning_rate": 0.00016,
+      "learning_rate_embeddings": 0.00016,
+      "loss": 3.8851,
+      "step": 8000
+    },
+    {
+      "epoch": 63.91,
+      "learning_rate": 0.00017,
+      "learning_rate_embeddings": 0.00017,
+      "loss": 3.796,
+      "step": 8500
+    },
+    {
+      "epoch": 67.67,
+      "learning_rate": 0.00017999999999999998,
+      "learning_rate_embeddings": 0.00017999999999999998,
+      "loss": 3.7154,
+      "step": 9000
+    },
+    {
+      "epoch": 71.43,
+      "learning_rate": 0.00019,
+      "learning_rate_embeddings": 0.00019,
+      "loss": 3.6434,
+      "step": 9500
+    },
+    {
+      "epoch": 75.19,
+      "learning_rate": 0.0002,
+      "learning_rate_embeddings": 0.0002,
+      "loss": 3.5794,
+      "step": 10000
+    },
+    {
+      "epoch": 75.19,
+      "eval_loss": 3.74064564704895,
+      "eval_runtime": 10.1022,
+      "eval_samples_per_second": 743.703,
+      "eval_steps_per_second": 5.84,
+      "step": 10000
+    },
+    {
+      "epoch": 78.95,
+      "learning_rate": 0.00021,
+      "learning_rate_embeddings": 0.00021,
+      "loss": 3.5056,
+      "step": 10500
+    },
+    {
+      "epoch": 82.71,
+      "learning_rate": 0.00022,
+      "learning_rate_embeddings": 0.00022,
+      "loss": 3.4434,
+      "step": 11000
+    },
+    {
+      "epoch": 86.47,
+      "learning_rate": 0.00023,
+      "learning_rate_embeddings": 0.00023,
+      "loss": 3.3841,
+      "step": 11500
+    },
+    {
+      "epoch": 90.23,
+      "learning_rate": 0.00024,
+      "learning_rate_embeddings": 0.00024,
+      "loss": 3.3289,
+      "step": 12000
+    },
+    {
+      "epoch": 93.98,
+      "learning_rate": 0.00025,
+      "learning_rate_embeddings": 0.00025,
+      "loss": 3.2738,
+      "step": 12500
+    },
+    {
+      "epoch": 93.98,
+      "eval_loss": 3.6198933124542236,
+      "eval_runtime": 10.0449,
+      "eval_samples_per_second": 747.942,
+      "eval_steps_per_second": 5.874,
+      "step": 12500
+    },
+    {
+      "epoch": 97.74,
+      "learning_rate": 0.00026000000000000003,
+      "learning_rate_embeddings": 0.00026000000000000003,
+      "loss": 3.2181,
+      "step": 13000
+    },
+    {
+      "epoch": 101.5,
+      "learning_rate": 0.00027,
+      "learning_rate_embeddings": 0.00027,
+      "loss": 3.1679,
+      "step": 13500
+    },
+    {
+      "epoch": 105.26,
+      "learning_rate": 0.00028000000000000003,
+      "learning_rate_embeddings": 0.00028000000000000003,
+      "loss": 3.1144,
+      "step": 14000
+    },
+    {
+      "epoch": 109.02,
+      "learning_rate": 0.00029,
+      "learning_rate_embeddings": 0.00029,
+      "loss": 3.067,
+      "step": 14500
+    },
+    {
+      "epoch": 112.78,
+      "learning_rate": 0.0003,
+      "learning_rate_embeddings": 0.0003,
+      "loss": 3.0177,
+      "step": 15000
+    },
+    {
+      "epoch": 112.78,
+      "eval_loss": 3.5627167224884033,
+      "eval_runtime": 10.0076,
+      "eval_samples_per_second": 750.731,
+      "eval_steps_per_second": 5.896,
+      "step": 15000
+    },
+    {
+      "epoch": 116.54,
+      "learning_rate": 0.00031,
+      "learning_rate_embeddings": 0.00031,
+      "loss": 2.9764,
+      "step": 15500
+    },
+    {
+      "epoch": 120.3,
+      "learning_rate": 0.00032,
+      "learning_rate_embeddings": 0.00032,
+      "loss": 2.9304,
+      "step": 16000
+    },
+    {
+      "epoch": 124.06,
+      "learning_rate": 0.00033,
+      "learning_rate_embeddings": 0.00033,
+      "loss": 2.8881,
+      "step": 16500
+    },
+    {
+      "epoch": 127.82,
+      "learning_rate": 0.00034,
+      "learning_rate_embeddings": 0.00034,
+      "loss": 2.8463,
+      "step": 17000
+    },
+    {
+      "epoch": 131.58,
+      "learning_rate": 0.00035,
+      "learning_rate_embeddings": 0.00035,
+      "loss": 2.8012,
+      "step": 17500
+    },
+    {
+      "epoch": 131.58,
+      "eval_loss": 3.5356414318084717,
+      "eval_runtime": 10.0661,
+      "eval_samples_per_second": 746.367,
+      "eval_steps_per_second": 5.861,
+      "step": 17500
+    }
+  ],
+  "max_steps": 250000,
+  "num_train_epochs": 1880,
+  "total_flos": 2.9264707584e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a8d9152003d509ba68460ab704ee1e323bbfa3f59b87fa0962fce2b0a2834e1
+size 2927